Title here
Summary here
Newsletter
Newsletter 22
10 February 2026AI learns to think in secret, Claude Opus spots evaluations, and Tesla robotaxis raise alarms
Newsletter 21
14 January 2026The Grok deepfake crisis, NIST's AI agent inquiry, and MATS Summer 2026
Newsletter 20
12 December 2025Genesis Mission, AI cyber-espionage, and alignment still unsolved
Newsletter 19
11 November 2025The end of OpenAI's nonprofit era, NIST probes Chinese models, and Anthropic finds introspection
Newsletter 18
14 October 2025Claude notices it's being tested, California moves on regulation, and competition worsens misalignment
Newsletter 17
12 August 2025Scheming evals under fire, Anthropic scans nuke queries, and Google faces safety accusations
Newsletter 16
10 August 2025GPT-5 reactions, the study that upended AI safety, and the Gemini CLI flaw
Newsletter 15
9 July 2025AI safety turns national-security-first, AlphaGenome, and robotics' "ChatGPT moment"
Newsletter 14
10 June 2025o3 resists shutdown commands, Anthropic builds for the military, and frontier models game reward systems
Newsletter 13
13 May 2025The Singapore Consensus, deceptive AI debate, and rolled-back safety measures
Newsletter 12
12 April 2025AI 2027, GATE, and the first AI Safety Talks event
Newsletter 11
11 March 2025Paris Summit regulation debates, new EU restrictions, and Eric Schmidt's $10M safety fund
Newsletter 10
11 February 2025DeepSeek shakes the race, U.S. AI policy shifts, and gradual disempowerment
Newsletter 9
14 January 2025Hinton's risk warning, o3's record, and Sora's launch
Newsletter 8
10 December 2024OpenAI o1, the Anthropic-Palantir partnership, and robot security flaws
Newsletter 7
12 November 2024AI Nobel prizes, White House security directive, and Claude controlling computers
Newsletter 6a
9 October 2024OpenAI's for-profit shift, executive departures, and the Council of Europe AI treaty
Newsletter 5
10 September 2024SB1047, China's changing AI safety stance, and AI Safety Camp 2025
Newsletter 4
13 August 2024EU AI Act, GPT-4o voice cloning, and foundation model governance
Newsletter 3
9 July 2024ARENA, AGI nationalization debate, and AI's impact on work
Newsletter 2
11 June 2024AI Safety Türkiye meet-up, fast grants, and Epoch AI's glimpse into the future
Newsletter 1
14 May 2024Introduction to AI Governance, Pivotal fellowship, and "What if Dario Amodei is right?"